Metode simpleks untuk regresi minmad (minimizing mean absolute deviations)

Yuliana , Lia (1999) Metode simpleks untuk regresi minmad (minimizing mean absolute deviations). Undergraduate thesis, FMIPA Undip.

[img]
Preview
PDF
16Kb
[img]
Preview
PDF
313Kb
[img]
Preview
PDF
234Kb
[img]
Preview
PDF
260Kb
[img]
Preview
PDF
376Kb
[img]PDF
Restricted to Repository staff only

788Kb
[img]
Preview
PDF
219Kb
[img]
Preview
PDF
209Kb
[img]PDF
Restricted to Repository staff only

1650Kb

Abstract

ABSTRAK Persamaan regresi biasa diselesaikan dengan meminimalkan jurntah kuadrat deviasi (metode kuadrat terkecil) dimana penerapannya barns berdasarkan asumsi kenormalan. Jika asumsi tidal( dipenuhi, maka dapat digunakan metode MINMAD. Tujuan penulisan adalah meneari estimasi parameter regresi dan deviasi mutlak minimal untuk membentuk regresi MINMAD. Metode yang digunakan adalab metode MINMAD yaitu metode yang meminimalkan rata-rata deviasi mutlak antara nilai pengamatan Yi dan nilai prediksi 1Tri dari pengamatan ke-i. Metode MINMAD disusun seperti program linier yaitu metode simpleks yang akan mengbasilkan parameter regress dan deviasi tnntiale. rn;n1 Mal, Se11:rit A:I (n - a. %At persamaan regresi MTN-MAD. Selanjutnya persamaan regresi MINMAD dibandingkan dengan persamaan regresi kuadrat terkecii ABSTRACT The Regression equation usually be found with minimizing the sum of squared deviations (least squares method) which use normal assumption. If these assumption be violated then it can be used MINMAD method. The research objects are to estimate the parameter of regressions and the minimized absolute deviations for MINMAD regressions. The MINMAD method is the minimizing mean absolute deviations between the observed and predicted values of Y1 the i-th observation. The MINMAD method cast be expressed bythe siraplex method of linear proga.atzilig to get the parameter of regressions and the minimized absolute deviations resulting MINMAD regression Finally, the MINMAD regression equation be compared with the least squared regression equation.

Item Type:Thesis (Undergraduate)
Subjects:Q Science > QA Mathematics
Divisions:Faculty of Science and Mathematics > Department of Mathematics
ID Code:31667
Deposited By:Mr UPT Perpus 2
Deposited On:24 Nov 2011 08:56
Last Modified:24 Nov 2011 08:56

Repository Staff Only: item control page